question

Ankur A3 avatar image
0 Likes"
Ankur A3 asked Jordan Johnson commented

How to check the model deadlock time while using optimizer?

Hi Team,

I am collecting results using optimizer for my model. But for 1 scenario, it is not giving results as expected and showing 0 value. Since model run time is around 50 hrs. it is little difficult to keep eye on it while running and checking it in 3D.

Is there any way we can get the model time when model is reaching in deadlock condition due to which it is not producing results?

or

If we can cover this kind of scenario in condition and print the model time at that moment?

Thank you!

FlexSim 23.1.0
optimizerdeadlock time
5 |100000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 12 attachments (including images) can be used with a maximum of 23.8 MiB each and 47.7 MiB total.

1 Answer

Jordan Johnson avatar image
0 Likes"
Jordan Johnson answered Jordan Johnson commented

I think the easiest way would be to add a Performance Measure that just reports the time:

Model.time

Then you could check that value for the replications that had a deadlock issue.

· 3
5 |100000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 12 attachments (including images) can be used with a maximum of 23.8 MiB each and 47.7 MiB total.

Ankur A3 avatar image Ankur A3 commented ·
Thank you! @Jordan Johnson

It makes sense.

But if deadlock condition is not happening and we want to dig down the reason it is giving 0 results for 1 scenario. What would you suggest?

I tried using debugger option while writing code, but it doesn't happen in optimizer.

Thank you!


0 Likes 0 ·
Jason Lightfoot avatar image Jason Lightfoot ♦♦ Ankur A3 commented ·

Apply that trials parameters to the model and run it interactively to see what's happening. You can also save the state of the models and open them for clues as to what might have happened.

0 Likes 0 ·
Jordan Johnson avatar image Jordan Johnson ♦♦ Ankur A3 commented ·
I agree with @Jason Lightfoot , you should run those scenarios interactively. If you click on a point in the optimizer chart, you should get an entry for that point in the legend, and you can click the test-tube button to apply those exact parameters, stop time and warmup time to your model. If running your model interactively doesn't have any problems, but running your model in the experimenter or optimizer does, then you have repeatability issues. Try running your model on FF to replicate how child processes run.
0 Likes 0 ·